Cell Phone and Brain Cancer

Motivation: Yesterday, I was talking on my new smartphone, and my ear grew warm.  The warmth spurred thoughts about a number of recent Google News posts warning about the dangers of cell phone use.  Do cell phones cause brain cancer?

Paper: Meta-analysis of long-term mobile phone use and the association with brain tumours.  Hardell, L. et al. Int. J. of Onc. (2008) 32: 1097-1103.

Method: Review of prospective and case-control studies that studied association of intracranial neoplasms and cell phone usage.

Results:
Glioma: In the meta-analysis using six case-control studies, the odds ratio of developing a glioma in the ipsilateral side to cell phone usage was 2.0 (CI: 1.2-3.4).  The odds ratio in the contralateral side was 1.1 (CI: 0.6-2.0).  The largest included study was a multi-national case-control study that studied cell-phone users, who had used cell phones for greater than ten years.  In the study, there were 77 cases of gliomas and 117 controls.  The odds ratio for gliomas in the side ipsilateral to cell phone use was 1.4 (CI: 1.01-1.9).  The OR for gliomas in the contralateral side was 1.0 (CI: 0.7-1.4).

Acoustic Neuroma: Nine case-control studies have examined this issue.  The meta-analysis yielded an odds ratio of 0.9 (CI: 0.7-1.1).  Of the nine studies, eight had not found any association.  When examining only the three studies looking at cell-phone users who had used cell phones for greater than 10 years, the odds ratio for acoustic neuroma in the ipsilateral side to use was 2.4 (CI: 1.1-5.3) and contralateral side was 1.2 (CI: 0.7-2.2).  Two of the three studies included in this analysis reached statistical significance for ipsilateral acoustic neuroma.

Meningioma: There were seven case-control studies studying this issue.  The meta-analysis gave an odds-ratio of 0.8 (CI: 0.7-0.99).  In the four case-control studies examining more than 10 year cell-phone users, the odds ratio for meningioma in the ipsilateral side was 1.7 (CI: 0.99-3.1) and in the contralateral side was 1.0 (CI: 0.3-3.1).  None of the four case-control studies reached statistical significance.

Conclusion: The association between cell-phone use and neoplasms is obviously of great public concern given the exploding number of global users.  The meta-analysis primarily found that cell-phone users of more than ten years had a greater likelihood of developing gliomas and acoustic neuromas in the side ipsilateral to cell phone use.  The study passes common sense tests because there appears to be greater association with length of use and site of use.  On the other hand, there are a couple of caveats.  The first is that this meta-analysis primarily consisted of case-control studies which are weak in proving causality.  People who use cell-phones, for instance, are also more likely to use other gadgets which emit radio-frequency or live in cities, where citizens may also be exposed to other toxins.  The second caveat is that gliomas are as such infrequent events, an odds-ratio of 2.0 may be scary but may not be very clinically significant (two times the odds of a rare event is not that frequent).  On the flip side, if the risk of neoplasm is dose dependent, the rates of neoplasms after multiple years may be significantly high.  For now, given our uncertain knowledge, I am buying a Bluetooth headset and getting a holster to separate the cell-phone from body contact!